Page 114 - Read Online

Page 6 of 14 Chia et al. Hepatoma Res 2019;5:9 I http://dx.doi.org/10.20517/2394-5079.2018.112

As a HCC diagnostic biomarker, GPC3 outperformed AFP in a number of independent studies.
[74]
Tangkijvanich et al. reported sensitivity of 56% for GPC3 while AFP stood at 33%. Interestingly, GPC3
overexpression did not correlate with AFP level, tumour size or stage of HCC but was significantly associated
with the presence of viral hepatitis markers. A meta-analysis of 19 studies reported the superior sensitivity
for GPC3 (pooled sensitivity, 55%; pooled specificity, 84%) over AFP (pooled sensitivity, 35%; pooled
[75]
specificity, 88%) . Notably, the specificity for GPC3 was significantly higher if the analysis focused only on
early HCC (pooled sensitivity 97%).

The potential of GPC3 goes well beyond being a diagnostic biomarker. GPC3 is an oncofetal antigen, a
protein that is predominantly expressed in cancer and during fetal development. Murine model injected
with GPC3 transgenic colon cell line showed that GPC3 was able to elicit T-cell mediated tumour rejection
[76]
[77]
without autoimmunity . Similar results were reported using highly metastatic mouse melanoma .
+
+
Importantly, the anti-tumour effects appeared to be mediated by both CD4 and CD8 T cells, which are
+
essential for optimal anti-tumour response. CD4 T cells have a broad role in orchestrating host anti-tumour
responses, such as secreting cytokines to enhance cytotoxic T cell response, activating eosinophils and
tumouricidal macrophages and secreting granulocyte/macrophage colony-stimulating factor. These results
encourage development of immunotherapy targeting against the tumor-specific GPC3 isoform.
A phase I clinical study of a GPC3-derived peptide vaccine for 33 advanced HCC patients demonstrated
[78]
that the vaccine was well-tolerated . Most patients had only grade I and grade II side effects. Four patients
developed grade III hematological adverse events but were likely due to disease progression rather than the
vaccine. In terms of efficacy, 1 patient showed partial response while 19 patients had stable disease 2 months
after initiation of treatment. Given the favorable safety profile, it is rational to target GPC3 with other
therapies, such as adoptive cell transfer. Along this line of thinking, there have been several Phase 1 clinical
studies of chimeric antigen receptor (CAR)-T trials directed against the GPC3 antigen in HCC patients.

Recently, a phase I clinical trial of anti-GPC CAR-T cells was conducted on Chinese patients (n = 13) with
+
refractory or relapsed GPC3 HCC. The 3rd generation CAR-T was engineered with CD28, 4-1BB and
[79]
CD3ζ downstream signaling domains . No dose-limiting toxicities were identified and only one patient
[80]
experienced grade 3 fever . Without lymphodepletion, none of the five patients responded to the treatment
but with lymphodepletion, 4/6 (67%) clinical response was reported. The authors concluded that the anti-
GPC3 CAR-T treatment is safe and tolerable.

A bispecific T cell-redirecting antibody that binds both GPC3 and CD3 is also under active development
by Chugai Pharmaceutical (a Roche subsidiary). Early studies on animal models showed that the GPC3/
CD3 bispecific antibody showed anti-tumour efficacy against various GPC3-positive xenografts including
[81]
liver tumours . This bispecific antibody is now being investigated in a phase I clinical trial on patients with
GPC3 positive advanced solid tumours (NCT02748837).

Cadherin-17
Cadherin-17 (CDH17) is a calcium-dependent cell adhesion molecule that belongs to the 7D-cadherin
superfamily, characterized by the presence of 7 cadherin-like ectodomains followed by a short cytoplasmic
[82]
tail . It is normally present in fetal liver and gastrointestinal tract during embroyogenesis, hence the
name liver-intestinal cadherin (LI cadherin). It is a peptide transporter and plays an important role during
embroyonic gastrointestinal development [83,84] . CDH17 expression was reported in normal human colon,
intestine and pancreas but not normal liver and stomach [85-89] . However, the overexpression of CDH17 was
observed in HCC as well as breast, ductal pancreatic, colorectal and gastric cancers [90-93] . The upregulation
was associated with malignant transformation of these cancers. Knock-down of CDH17 by RNAi inhibited
[94]
proliferation of primary and metastatic HCC cell lines in vitro and in vivo . This anti-tumour effect
was likely due to inactivation of Wnt signaling pathway because CDH17-knockdown HCC tumours